We use cookies to ensure that we give you the best experience on our website. You can change your cookie settings at any time. Otherwise, we'll assume you're OK to continue.

Durham University

The Department of Philosophy


Publication details for Professor Nancy Cartwright

Cartwright, N. (2010). Hunting Causes and Using Them: Approaches in Philosophy and Economics. Analysis 70(2): 307-310.

Author(s) from Durham


Hunting Causes and Using Them: Approaches in Philosophy and Economics (HC&UT) is about notions of causality appropriate to the sciences, mostly generic causal claims (causal laws) and especially notions that connect causality with probability.1 Most of the work for the book is associated with the project ‘Causality: Metaphysics and Methods’. This project argued that metaphysics – our account of what causal laws are or what general causal claims say – should march hand-in-hand with our ways of establishing them. It should be apparent, given the kind of thing we think causality is, why our methods are good for finding it. If our metaphysics does not mesh with and underwrite the methods, we are willing to trust, we should be wary of both.

Many philosophers nowadays look for a single informative feature that characterizes causal laws. HC&UT argues instead for causal pluralism, for a large variety of kinds of causal laws as well as purposes for which we call scientific claims causal. Correlatively different methods for testing causal claims are suited to different kinds of causal laws. No one analysis is privileged and no methods are universally applicable. Much of the argument for pluralism is provided by authors of different accounts of causality, who provide intuitively plausible counter-examples to each other. Still, most of these accounts seem adequate for the kinds of examples the authors focus on. From the point of view of HC&UT, these examples involve different kinds of causal laws or set causality to different jobs, and the concomitant characterizing feature marks out this one kind of causal law.

Importantly for the argument, often we can specify what characteristics a system of laws should have in order for an account/method pair to be applicable. An example is James Woodward’s level invariance, which I see as a …